Eprom Load, Ram Load.... No run.

TJHJ

So yesterday this board a USB Proto board, was working fine, It was stored in my garage overnight, we have been seeing sub freezing tempatures here, and the garage is not well insulated. My 1st suspect was a too cold of cystal so I warmed it slightly using my hand. After turning on the heat in the garage I noticed some small condensatation, on the board, and apartus.

But I can reload the program to both ram and e-prom with out any errors. But it does not run at all....

Any ideas similar problems? Is the prop sensitive to cold weather? I saw the heat test on You tube, where they had one being cooked and it was fine....

Thanks for the help as always
TJ

Mike Green

The infamous PLL failure behaves the way you describe as well as a bad connection at the crystal.· The Prop can identify itself to the Prop Tool, download programs to RAM and EEPROM, but the programs don't run.· Try some kind of test program that uses only the RCFAST clock like an LED blink program.· Serial I/O won't work without a working crystal driven system clock.

If the RCFAST program works, try reseating the crystal after cleaning (wiping) under it in case there was some debris that got on the board.

Philldapill

You said it's a protoboard, and those crystals are nice and snug. I hate to say it, but Mike is probably right about the PLL issue if the RCFAST configuration works. A funeral might be in order for that chip.

grasshopper

I am running a liquid cooled prop and it gets below -10c; runs fine. Condensation does seem to build up on the cooler but never has it effected my propeller yet. Strange that yours wont run correctly. Try loading a different code perhaps you are running something faulty.

TJHJ

A few hours next to the heater and it works again....
Condensation? When it was dead, I put a scope on the crystal, and it was working. I connected the scope to the socket side to also check connections.
It would be nice to nail this down so I can prevent it in the future.
Grasshopper curiosity, what takes a liquid cooled prop, I cant ever seem to make one ever generate enough heat to effect it.
·
TJ
·

grasshopper

The liquid cooling is actually used for the prop, a mosfet and an AtoD converter. I am measuring to a great degree of detail and so this cooling keeps the temperature on all the important IC' s to a known value. This gives me the ability to keep some things constant, or at least keep the IC's cool so that I don't loose any significant bits.

Dr_Acula

The condensation could be a problem. Water is conductive so you could effectively have pins shorted. Plus you will get away with condensation for a few weeks or months but soon the pins may start oxidising. I pulled a board out of an outdoor setup recently that had only been outside for a year and the ICs had no pins left and had fallen off the board!

Can you put it in a waterproof case?

TJHJ

WaterProof case will be tough, and keeping it sealed, 26 connections going all over this thing... I picked up some silicon conformal coating, Ill let you know if it works out..

TJ

grasshopper

Dr_Acula (James Moxham) said...
The condensation could be a problem. Water is conductive so you could effectively have pins shorted.

Actually water is not conductive. Its the impurities in the water that is making it seem as though it is conducting. There is not enough valance electrons to conduct electricity (under normal conditions) thus it makes a fairly decent insulator.

Don't mean to be picky or not trying to offend, but some people may get misguided in that statement.

@TJHJ
I am using a conformal coating on my liquid cooled prop similar as to what you mentioned. It seems to be working very well. Its easy to apply and inexpensive.

Post Edited (grasshopper) : 1/1/2009 6:40:15 AM GMT

kwinn

TJ, having worked in a garage that was only heated when I was working there I am familiar with the kind of problems you are having. The solution came to me when my refrigerator compressor died. I bought a new one and put the old one in the garage to store my boards and tools in. The only change I made was to put a thermostat in parallel with the door light to keep it warm inside.

Oldbitcollector (Jeff)

@grasshopper

It escapes me how you are managing to generate heat from the Propeller.
I can see it from the mosfet, but the Propeller?

I asked Chip at the expo about whether the PropII would require some sort
of heatsink or cooling. He strongly indicated this would not
be the case as his design does not waste energy in the form of heat.
Kinda gives you some real insight about CPU design doesn't it?

OBC

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
New to the Propeller?

Check out: Protoboard Introduction , Propeller Cookbook 1.4 & Software Index
Updates to the Cookbook are now posted to: Propeller.warrantyvoid.us
Got an SD card connected? - PropDOS

grasshopper

Oldbitcollector said...
@grasshopper

It escapes me how you are managing to generate heat from the Propeller.
I can see it from the mosfet, but the Propeller?

I asked Chip at the expo about whether the PropII would require some sort
of heatsink or cooling. He strongly indicated this would not
be the case as his design does not waste energy in the form of heat.
Kinda gives you some real insight about CPU design doesn't it?

OBC

Sorry for misleading you.

To put it more clearly I have some Ic's in a close proximity to each other where some generate tremendous heat (especially the MosFet IC - 30Amps). The heat that is generated will begin to heat the surrounding parts quickly and thus I would loose my accuracy especially from the AtoD converter (16bits over 5volts). I overcome this by placing a liquid cooled heat exchanger over most of the circuit board that is populated with the propeller and other important Ic's. I am using the quad package that the propeller comes in - the size of the heat exchanger is under 1.75" square so it covers the parts well.

All I was stating earlier is that I am liquid cooling the propeller (and other IC's) and I have never had it fail yet. I have seen condensation and cooled it below -10c with no problems.